— Dual-path abliteration targeting LFM2.5 hybrid architecture (18 conv + 6 attention layers) — 50 modules modified across all 24 layers (conv + attention paths) — Refusal direction: harmful — harmless — Per-module weights: attn=4.0, conv=3.0, ffn=2.5, in_proj=2.0 — 51 prompts for precise direction estimation LFM2.5 is a new family of hybrid models designed for on-device deployment. It builds on the LFM2 architecture with extended pre-training and reinforcement learning. — On-device personal assistant: Designed to power real-life applications, chaining tool calls, and following complex instructions on all devices. — Compressed performance: Competitive with much larger dense and MoE models on instruction following and agentic tasks. — Unmatched throughput: Fastest in its size class on both CPU and GPU inference, with day-one support for llama.cpp, MLX, vLLM, and SGLang. — Model size: 8B params — Tensor type: F32 / BF16 — Architecture: Hybrid (18 conv + 6 attention layers) — Paper: LFM2 Technical Report
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: blockblockblock
Теги: lfm2_moe, lfm2.5, liquid, conversational, uncensored, abliterated, endpoints_compatible
Лайков: 4 | Загрузок: 71
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
